AVT, in both variants, AVT-R and AVT-O, is an elegant treatment and has attracted the attention of many units that want to optimize results. The use of only volatile substances allows blowdown to be minimized, and apparently brings easier control.
The first requirement is that the makeup water has the required purity. In AVT treatments, blowdown is minimized or even eliminated, and for this reason the removal of impurities coming from the makeup water and from the condensate return is greatly compromised. With no exit route, any contaminant that enters through the makeup stays in the system until it is physically removed, usually already deposited or having already corroded some surface.
The water quality required is full demineralization. Typical values cited in the literature recommend specific conductivity below 0.1 µS/cm and total silica below 10 ppb. This level is normally only reached in systems with reverse osmosis followed by polishing, or mixed-bed ion exchange. Adopting AVT with merely softened water, or with an undersized or poorly maintained demineralization system, is setting up a program that does not hold up in practice.
Another restriction that is also not taken into account is the low buffering capacity. This low capacity means there is no chemical reserve to cushion variations. These variations can lead to the onset of a corrosive process.
Low buffering changes the way analytical monitoring must be done. A deviation of a few hours that goes undetected is already enough time to change the condition of the protective film, whether it is Cu₂O in AVT(R) or the double oxide layer in AVT(O). Continuous instrumentation for monitoring and control of pH, ORP and dissolved oxygen, with an alarm configured for deviations, stops being a desirable feature and becomes a condition for the safe operation of the program.
A third recurring mistake is not adjusting the program during startup, shutdown or layup of the equipment. In AVT(O), for example, the controlled residual oxygen condition presupposes normal operation and does not automatically cover what happens when the unit is shut down and the surface is exposed to air at a higher concentration than expected in operation. Treating startup and shutdown as an extension of the normal operating regime, without a specific procedure, opens a corrosion window that routine monitoring will not catch in time.
The fourth mistake is a condenser contamination that is not detected quickly. In AVT(R) cycles, an unexpected ingress of contaminated cooling water not only affects the purity of the feedwater, but can also shift the ORP out of the range that protects copper.
In AVT(O) cycles, the same leak can abruptly change the dissolved oxygen residual, pushing the system out of the range that sustains the protective film and possibly turning into localized corrosion. In both cases, the time between the contamination entering and its detection is what determines the size of the damage.
Another mistake is not reassessing the program after a physical change in the cycle. A retrofit that removes copper-alloy low-pressure heaters makes the system, from a metallurgical standpoint, eligible to migrate from AVT(R) to AVT(O). Keeping the old program out of inertia, without reassessing the current metallurgy of the cycle, means giving up the gain of a higher pH range and the more robust protection that AVT(O) offers for fully ferrous systems, discussed in the previous post of this series.
The reverse also happens. A modification that introduces a copper-alloy heat exchanger into a previously all-ferrous cycle requires the opposite path, lowering the pH range to the limit compatible with copper. In both directions, the right decision is a technical one, that is, review the program with every change in metallurgy.
The pattern common to all these mistakes is treating AVT as a simpler program because it has no dissolved solids, when in practice it demands more rigor, since it is more subject to variation. The chemical simplicity of AVT implies greater care with feedwater quality. Continuous monitoring and operating discipline during regime transitions are requirements. Understanding this at the program decision stage avoids a good share of the problems that normally only appear later.
In the next post of this series, we turn to pH control in AVT, the comparison between ammonia and neutralizing amines as the volatile alkalizing agent, and what changes in the choice between them.
Has your plant assessed whether the makeup water quality truly supports the AVT program adopted, or does it still operate with a safety margin that does not exist in practice? Talk to us for a technical assessment.
References
- IAPWS TGD3-10(2015): "Volatile treatments for the steam-water circuits of fossil and combined cycle/HRSG power plants". Available at: iapws.org.
- ScienceDirect Topics. All-Volatile Treatment - an overview. Available at: sciencedirect.com.
- ChemTreat. The Challenges of Industrial Boiler Water Treatment. Available at: chemtreat.com.
- US Patent 4,827,959 A. Monitoring and controlling AVT (all volatile treatment) and other treatment programs for high pressure boilers via the conductivity control method. Available at: image-ppubs.uspto.gov.